1. Field of the Invention
The invention relates to twisted-pair planar cables, and particularly to the twisted-pair cable having 90 degrees phase/node offset between every adjacent two pairs cables.
2. The Related Art
The twisted pair electrical cable has long been known to posses much better electrical characteristics than the purely parallel lay wires or cable, e.g., flat cable, wherein each pair of insulated conductors are typically twisted together along the length of the cable. The geometry of the twisted pairs is known to produce the desirable electrical characteristics of low crosstalk and high immunity from interference from external magnetic fields. FIG. 1 which is the main drawing of U.S. Pat. No. 4,486,619, shows the first generation conventional twisted pair electrical cable.
FIG. 2 shows the second generation conventional electrical cable wherein every adjacent two pairs of conductors are twisted in a reverse manner with each other. For example, in FIG. 2 the first upper pair 1 are twisted along its length clockwise from a right side viewpoint position, and the second upper pair 2 are twisted along its length counterclockwise from the same right side viewpoint position. And the third upper pair 3 are again twisted along its length clockwise from the same right side viewpoint position.
This alteration is to try to further counterpoise the Electro-magnetic effects thereamoung, so as to expect to reduce the induce the signals(noises) of one pair of conductors due to its adjacent pairs of conductors. In other words, the twisted pair cable is adopted to reduce crosstalk noise in the differential signal applications because the wires in the pair are close to each other to expect to have the common-mode induced signals from the noise source and such common-mode induced signals will be counterpoised in the differential applications. Additionally, the overall Electro-magnetic filed is expected to be reduced due to the opposite polarity between the neighboring loops in the pair.
Anyhow, the results are not as good as expected. The reason is that this concept may be workable when the subject pair of conductors are far distant from the adjacent pair of conductors. Under that situation, the distance between the adjacent pairs of conductors is much larger than the distance between the two internal conductors within each pair of conductors, and thus the influence applied to the pair of conductors can be regarded as from a single source of the adjacent pair of conductors.
In fact, oppositely the pairs of conductors are closed side by side arranged with each other. The distance between the adjacent pairs of conductors are generally close to the distance between the two internal conductors within each pair of conductors. Under this situation, the influence of the subject pair of conductors should be measured by the individual conductors of the adjacent pair of conductors. In other words, when the ratio of the distance between the twisted pairs to the distance between the internal conductors(wires) in the each twisted pair becomes smaller, the difference of the induced noises of the two internal conductors of the twisted pair will become larger, and can not be treaded as common-mode induced signals. Thus, a net crosstalk noise occurs. The reasons are as below.
As shown in FIGS. 3(A) and 3(B), for the uniform twisted pair cable of FIG. 2, the far end crosstalk may be unstably either accumulated or cancelled in the individual period depending upon the data pattern of the current in the neighbor pairs. As understood, the induced noises of the conductor of each pair are mostly effected only by the two closest neighbor pairs by two sides of the subject pair, and thus the discussion is among these three pairs. Anyhow, it should be understood that each conductor of the subject pair can be a victim one which is influenced by the conductor of the adjacent pair while simultaneously it can also be a driving one influencing the other conductor of the adjacent one.
In FIG. 3(A), the upper or driving pair 50 include the conductor 52 assuming with xe2x88x921 V thereof and the conductor 54 assuming with +1 V thereof, and the lower or driving pair 60 include the conductor 62 assuming with xe2x88x921 V thereof and the conductor 64 assuming with +1 V, to be commonly positioned by two sides of the middle or victim pair 70 include the conductors 72 and 74. For the conductor 72, the current of the victim conductor 72 will carries an induced signal(noise) occurring around the point 72A due to inducement to the driving conductor 54 and sequentially carries another induced signal occurring around the point 72B due to inducement to the driving conductor 64. Understandably, the similar induced signals will occur around every closer positions between the victim conductor 72 and the corresponding driving conductors 54, 64, and be accumulated along the whole length of the victim conductor 72. Undoubtedly, the far end P of the cable will have a high crosstalk thereof. The other conductor 74 follows the same format.
Differently, if the data pattern of the current is arranged as in FIG. 3(B) wherein the conductor 52 assuming with +1 V instead of xe2x88x921 V in FIG. 3(A), and the conductor 54 assuming with xe2x88x921 V instead of +1 V FIG. 3(A), the results will be totally different. Under this situation, for the conductor 72, the current of the victim conductor 72 will carry the induced signal around the point 72A which is generated by the driving conductor 54, and further carry the induced signal around the point 72B which is generated by the driving conductor 64, while the former induced signal and the latter induced signal will be mutually eliminated with each other because of the opposite polarities of the driving conductors 54 and 64. Understandably, the far end P of the conductor 72 of the cable has a lower crosstalk. The other conductor 74 follows the same format.
Understandably and unfortunately, the data pattern of the current in each conductor is of an indirect current and thus not controllable. +1 V and xe2x88x921 V are interchanged with each other several times in the conductor within each second whereby the data pattern in each conductor is randomly different at each specific moment. It is impossible to arrange the whole system as in a format shown in FIG. 3 consistently and permanently. Therefore, the crosstalk around the far end of the cable is uncontrollably too high to be acceptable.
Recently, some attempts have been taken to reduce these induced signals from different approaches, wherein irregularities arrangement has applied to the conductors either within each pair of conductors or among the adjacent pairs of conductors for lowering crosstalk, insertion loss or return loss, for example, U.S. Pat. Nos. 3,736,366 and 5,767,441. Anyhow, the random, changeable, or irregular arrangement of the pairs of conductors is not a scientific way to either design or manufacture the corresponding components where the designer has no strong confidence in quality consistence or precisely manufacturing. Therefore, it is desired to have a scientific, systematic, reliable and predicable arrangement in the twisted pair cable for efficiently reducing the induced noises thereof.
According to an aspect of the invention, an electrical cable includes a plurality of twisted pairs of conductors thereof. The pair of conductors are twisted with some degrees axially offset relative to the adjacent twisted pair of conductors, whereby for each victim conductor of each pair of conductors which is measured for each twisting cycle by 0 through 360 degrees, the first section within the 0-90 degrees range is influenced by one first conductor of one adjacent first pair of driving conductors while simultaneously the second section within the 90-180 degrees range is influenced by the other second conductor of the same one adjacent first pair of driving conductors under the condition that the first conductor and the second conductor are of opposite polarities, thus the combined induced signals(noises) of the victim conductor is close to zero. Similarly, for the same victim conductor of the same pair of conductors, the third section within the 180-270 degrees range is influenced by one third conductor of the other adjacent second pair of driving conductors while simultaneously the fourth section within the 270-360 degrees range is influenced by the other fourth conductor of the same other adjacent second pair of driving conductors under the condition that the third conductor and the fourth conductor are of opposite polarities, thus the combined induced signals(noises) of the victim is close to zero. Accordingly, for each victim conductor within each twisting cycle of 0-360 degrees range, the compound induced signal is itself eliminated as low as zero. Thus, the far end of the cable through several twisting, still owns the very low value of the induced signal.